Remove Advertisements

Heroism timing

Warning: Theorycraft inside.

Moderators: Fridmarr, Worldie, Aergis, theckhd

Postby Varmin » Wed Mar 04, 2009 4:19 pm

In reply to rotation changes in execute range:

The only way this would change the equation, is if not having bloodlust in execute range would prohibit you from changing your rotation. As long as bloodlust is the same haste increase before or after execute range, rotation changes won't affect this at all.
Varmin
 
Posts: 172
Joined: Sun Jan 18, 2009 8:57 pm

Postby cordelia » Wed Mar 04, 2009 4:21 pm

Markoh wrote:
holythief wrote:Which all adds up the fact that heroism should be triggered in the execute range for every group.


THere have been a couple posts in different places on this recently. Basically the problem with lusting at the end of a fight is that it is almost impossible for everyone to have cooldowns ready to use without wasting them (aka sitting on cooldowns b/c your saving for lust). I don't have a link off hand but basically what the math needs to come down to is the dps gain by lusting in execute range vs lusting 10-15 seconds in the fight with everyone blowing full cooldowns.


This is important, and I touched on it in my rules, but didn't go into details. Honestly, I think the two best times to maximize heroism returns are probably either:

(a) right after the dps gets settled at the start of the fight, or

(b) around the 45 second mark (assuming the raid knows it will be blown 50 seconds in).

Why? Obviously on-use abilities and trinkets can be controlled to overlap heroism. However, currently many dps are using trinkets very similar to the Mirror of Truth, which has ~45 second internal CD. You want maximum overlap with the greatest # of possible random trinket procs in the raid. (a) seems to guarantee that you will get almost every significant CD and trinket proc used. (b) comes closest to guaranteeing this after (a).

Pros to (a):
- guarantee of every significant trinket proc
- guarantee of every significant on-use CD
- guarantees maximum uptime of on-use CD's, eg. wings can be blown as many times as possible in this situation, whereas saving wings for 50 seconds in might lose a wings use for ret pallies.
- guarantees all 25 raiders are alive

Pros to (b):
- no dps loss due to certain classes requiring "setup" debuffs.
- no dps loss due to long CD debuffs and auto-refresh debuffs being cast.
- fewer threat issues


IMO, if threat is not an issue, (a) is best. Otherwise, (b).

And yes, this assumes that there are no time advantages to getting heroism in certain phases. It also assumes you can't guarantee raider uptime if there is in fact a disproportionate gain from heroism due to Execute range mechanics.
-
User avatar
cordelia
 
Posts: 630
Joined: Tue Oct 09, 2007 3:33 am

Postby Mordinm » Wed Mar 04, 2009 5:04 pm

Varmin wrote:In reply to rotation changes in execute range:

The only way this would change the equation, is if not having bloodlust in execute range would prohibit you from changing your rotation. As long as bloodlust is the same haste increase before or after execute range, rotation changes won't affect this at all.


This is not correct at all. Bloodlust is not a flat % damage increase it is a haste increase. The math assumes that bloodlust will increase your dps by a given percentage and that percentage does not change based on when you pop lust. If the rotation change changes the value of haste then one of the core assumptions of the model is wrong.

Warriors move from a rotation limited by cool downs to one limited only by rage gen. As such haste becomes a much more valued stat because increasing white damage by 30% increases yellow damage by 30% as well. Above 20% increasing white damage by 30% will not lead to a 30% increase in yellow damage due to the optimum rage uses having cool downs that are not affected by haste.

Ret pallies on the other simply add another cool down to their rotations so haste becomes less of a buff. Perhaps it would be more accurate to say that lust would provide the same numerical dps buff to a ret pally, say 200 dps, but as the ret pally would be doing more dps thanks to hammer throw the percentage dps buff would drop.
Mordinm
 
Posts: 307
Joined: Tue Apr 22, 2008 10:58 am

Postby trellian » Thu Mar 05, 2009 1:28 am

Markoh wrote:I don't have a link off hand but basically what the math needs to come down to is the dps gain by lusting in execute range vs lusting 10-15 seconds in the fight with everyone blowing full cooldowns.


This is only true if those cooldowns scale with haste. Most cooldowns don't (just a static +SD or +AP), and some cooldowns it would actually lower overall DPS (think of +haste trinkets).
Saving cooldowns for during heroism is only useful if you need the heroism to get through a phase quickly (Maexxna enrage for example). In all other cases you should just use cooldowns whenever they are available, whether you have heroism or not; as the percentage increase of using heroism is the same whether you use cooldowns or not (in some rare cases using heroism together with cooldowns the increase is lower).

On progression you generally want to heroism early, just so you can see more phases of the fight and learn the encounter better.


The only time where heroism combined with cooldowns results in a more increase in (total average bosskill) dps is if your rotation during those cooldowns change and the abilities you use during cooldowns scale with haste better then the abilities you use without cooldowns. And this is (afaik) not the case with any class. Saving cooldowns for during heroism is thus a bad idea unless you need to blow through a phase as fast as you can. You will see a more increase in (total average bosskill) dps is you use cooldowns as soon as possible and as often as possible.


## in this whole post you can substitute cooldowns with procs as they are true aswell.
User avatar
trellian
 
Posts: 671
Joined: Wed Jul 16, 2008 3:02 am
Location: Silvermoon, EU

Postby Kalakaua » Thu Mar 05, 2009 1:53 am

The math is correct, but the assumptions are wrong. If you assume that heroism increases your dps by 30% at the beginning of the fight, it will also increase your dps by 30% at the end of the fight. This is not true for some classes like fury warriors.

When the boss is at full health, their rage dump is heroic strike which is limited by their weapon swing. At less than 20% health, they can dump their rage into execute which is an instant attack.

If a warrior does 1000 dps at the beginning and 1200 dps at the end, the assumption is made that the 30% increase from heroism will let the warrior either do 1300dps at the beginning or 1560 dps at the end. In reality, if the warrior gets a 30% boost at the beginning from heroism, he will likely get a 40% boost by using it at the end and be doing more like 1680dps.
Kalakaua
 
Posts: 212
Joined: Fri May 16, 2008 11:05 am

Postby Varmin » Thu Mar 05, 2009 2:05 am

Kalakaua wrote:The math is correct, but the assumptions are wrong. If you assume that heroism increases your dps by 30% at the beginning of the fight, it will also increase your dps by 30% at the end of the fight. This is not true for some classes like fury warriors.

When the boss is at full health, their rage dump is heroic strike which is limited by their weapon swing. At less than 20% health, they can dump their rage into execute which is an instant attack.

If a warrior does 1000 dps at the beginning and 1200 dps at the end, the assumption is made that the 30% increase from heroism will let the warrior either do 1300dps at the beginning or 1560 dps at the end. In reality, if the warrior gets a 30% boost at the beginning from heroism, he will likely get a 40% boost by using it at the end and be doing more like 1680dps.


This is also in response to the previous poster as well.

This is the only example where bloodlust in execute range would give more dps. However, I haven't ever seen the math of where bloodlust would be a greater % increase in DPS in execute range for a warrior. These just seems like a very bad assumption from people noticing that warriors gain alot of DPS during execute range. Even if a warrior gains 10k DPS from execute range, if he only gains the same % of DPS gain, there is no change of when you bloodlust.
Varmin
 
Posts: 172
Joined: Sun Jan 18, 2009 8:57 pm

Postby Varmin » Thu Mar 05, 2009 2:20 am

cordelia wrote:
Pros to (a):
- guarantee of every significant trinket proc
- guarantee of every significant on-use CD
- guarantees maximum uptime of on-use CD's, eg. wings can be blown as many times as possible in this situation, whereas saving wings for 50 seconds in might lose a wings use for ret pallies.
- guarantees all 25 raiders are alive



This is false. Only haste affects will benefit from bloodlust.

Take wings for example:

Wings will always give 20% more damage with or without bloodlust. This is the same for flat AP increases. (You don't magically gain more AP from a proc just because you have bloodlust).

Haste affects are multiplicative. A 20% haste cooldown will give you 20% extra haste without bloodlust or other affects, but will give you 26% haste during bloodlust.
Varmin
 
Posts: 172
Joined: Sun Jan 18, 2009 8:57 pm

Postby cordelia » Thu Mar 05, 2009 4:05 am

Varmin wrote:
cordelia wrote:
Pros to (a):
- guarantee of every significant trinket proc
- guarantee of every significant on-use CD
- guarantees maximum uptime of on-use CD's, eg. wings can be blown as many times as possible in this situation, whereas saving wings for 50 seconds in might lose a wings use for ret pallies.
- guarantees all 25 raiders are alive



This is false. Only haste affects will benefit from bloodlust.

Take wings for example:

Wings will always give 20% more damage with or without bloodlust. This is the same for flat AP increases. (You don't magically gain more AP from a proc just because you have bloodlust).

Haste affects are multiplicative. A 20% haste cooldown will give you 20% extra haste without bloodlust or other affects, but will give you 26% haste during bloodlust.


Huh? what is false?

TRUE: Hitting bloodlust at the start guarantees almost every major trinket will go off because almost every proc based trinket will proc within the first 40 seconds of a fight.

TRUE: Hitting bloodlust at the start allows raiders to immediately use on-use trinkets.

TRUE: Hitting bloodlust at the start allows maximum uptime of on-use CD's such as wings. In a 2:30 second fight, wings can be blown twice, but only if it is blown within the first 10 seconds of the fight. If you wait for 50 seconds, it can only be used once.

TRUE: Hitting bloodlust at the start guarantees that all 25 raiders are alive. Umm... this is kinda obvious.


So, err, exactly what is false, and why? Because your reasoning makes no sense to me and I don't even know what you're addressing.
User avatar
cordelia
 
Posts: 630
Joined: Tue Oct 09, 2007 3:33 am

Postby Varmin » Thu Mar 05, 2009 4:09 am

Bloodlust doesn't need to be timed with any non haste trinkets or cooldowns. DPS should be using these trinkets/cooldowns at around 15s into the fight anyway. There is no point for them to wait 50s for bloodlust.
Varmin
 
Posts: 172
Joined: Sun Jan 18, 2009 8:57 pm

Postby cordelia » Thu Mar 05, 2009 5:44 am

Varmin wrote:Bloodlust doesn't need to be timed with any non haste trinkets or cooldowns. DPS should be using these trinkets/cooldowns at around 15s into the fight anyway. There is no point for them to wait 50s for bloodlust.


Let's assume Bloodlust only affects white damage for a ret pally. Let's say white damage is 800 DPS. Wings will increase white damage by 160 DPS. Heroism will increase white damage by 160 DPS. Wings and Heroism together will increase white damage by 352. Using them separately grants 320. Thus, overlapping Wings with Bloodlust grants a dps increase of 32 for the overlap time.

So there is an added benefit of blowing Bloodlust while you blow wings during the first 40 seconds of combat. Perhaps I should restate it as guarantees uptime with on-use CD's longer than 1 min.
User avatar
cordelia
 
Posts: 630
Joined: Tue Oct 09, 2007 3:33 am

Postby trellian » Thu Mar 05, 2009 5:46 am

FALSE: assuming haste effects your dps more when using cooldowns then when not using cooldowns

Varmin wrote:Bloodlust doesn't need to be timed with any non haste trinkets or cooldowns. DPS should be using these trinkets/cooldowns at around 15s into the fight anyway. There is no point for them to wait 50s for bloodlust.


This...
User avatar
trellian
 
Posts: 671
Joined: Wed Jul 16, 2008 3:02 am
Location: Silvermoon, EU

Postby amh » Thu Mar 05, 2009 6:08 am

Varmin wrote:Take wings for example:

Wings will always give 20% more damage with or without bloodlust. This is the same for flat AP increases. (You don't magically gain more AP from a proc just because you have bloodlust).


Not quite sure I agree here. Possibly a slight derail since I´m not factoring in execute-range. Anyhoo:

Wings will give 20% more damage with or without heroism, yes. The damage increase from specials would be the same with / without heroism, since haste doesn´t really help much and we´ve got no spammable abilities.

However, the damage increase from white damage and on-use effects is higher, simply because you perform more attacks. Don´t dwell too much on the weapon speed, it´s probably inaccurate, but you get the idea.

40 second span, weapon-speed 3.0 raid-buffed
Without heroism you perform 13,3 normal attacks, 6,6 of which are affected by wings and, let´s say this trinket

With heroism you perform 18-19 normal attacks, 9,5 of which are affected by wings and popped trinket. It´s not much, but you do gain more damage from your wings and on-use effects with heroism than without.

Or am I missing something obvious?
I used to play a paladin.
User avatar
amh
Maintankadonor
 
Posts: 3012
Joined: Fri May 02, 2008 6:25 am
Location: Oh hi

Postby Elsie » Thu Mar 05, 2009 6:39 am

Mm...

I was a bit hazy last night, but I seem to recall it went like this:
We have a defined X, X*Y, Z, and W.

The proof implies Z = (X*Y-X)*W and later states the Damage Done = X*Z*W where Z is the marginal DPS gain from Lust during W.

However, If X is original DPS, and Z is marginal dps gain, Damage Done = (X+Z)*W. We note that X+Z = X*Y.

We also see can note that X*Y = X*(1+r) where r is the rate of DPS increase from lust. Thus, r = Z. Therefore, X+Z = X(1+Z). This implies that X+Z =X +XZ.

Now, we refer to Z = (X*Y - X)*W. This implies X*Y-(Z/W) = X
=> X*(1+Z) - Z/W) = X using X*Y = X*(1+r) = X(1+Z)
=> X +XZ - (Z/W) = X
=> XZ - (Z/W) = 0 by adding -X = -X
=> XZ - (Z/W) = 0
=> Z*X - Z (1/W) = 0
=> Z *(X - 1/W) = 0
=> Z = 0.

Thus, by application of the proof's assumptions, we see that bloodlust grants no DPS. This is a contradiction.
User avatar
Elsie
 
Posts: 3819
Joined: Sat Jan 12, 2008 11:12 pm

Postby Mordinm » Thu Mar 05, 2009 6:54 am

Varmin wrote:Bloodlust doesn't need to be timed with any non haste trinkets or cooldowns. DPS should be using these trinkets/cooldowns at around 15s into the fight anyway. There is no point for them to wait 50s for bloodlust.


You know when you say crap like this it's nice if you give some sort of reasoning rather then just asserting that the other person is wrong.

That said you are absolutely completely wrong see Amh's post for the reason why. It been explained several times in this thread that the proof in the OP only applies for dps changes that are dependent on boss health.
Mordinm
 
Posts: 307
Joined: Tue Apr 22, 2008 10:58 am

Postby Mordinm » Thu Mar 05, 2009 7:08 am

Elsie wrote:Mm...

I was a bit hazy last night, but I seem to recall it went like this:
We have a defined X, X*Y, Z, and W.

The proof implies Z = (X*Y-X)*W and later states the Damage Done = X*Z*W where Z is the marginal DPS gain from Lust during W.

However, If X is original DPS, and Z is marginal dps gain, Damage Done = (X+Z)*W. We note that X+Z = X*Y.

We also see can note that X*Y = X*(1+r) where r is the rate of DPS increase from lust. Thus, r = Z. Therefore, X+Z = X(1+Z). This implies that X+Z =X +XZ.

Now, we refer to Z = (X*Y - X)*W. This implies X*Y-(Z/W) = X
=> X*(1+Z) - Z/W) = X using X*Y = X*(1+r) = X(1+Z)
=> X +XZ - (Z/W) = X
=> XZ - (Z/W) = 0 by adding -X = -X
=> XZ - (Z/W) = 0
=> Z*X - Z (1/W) = 0
=> Z *(X - 1/W) = 0
=> Z = 0.

Thus, by application of the proof's assumptions, we see that bloodlust grants no DPS. This is a contradiction.


Z is the percentage dps gain. Damage done while lusted = Z*X*W not (Z+X)*W
Mordinm
 
Posts: 307
Joined: Tue Apr 22, 2008 10:58 am

PreviousNext

Return to Advanced Theorycraft and Calculations

Who is online

Users browsing this forum: No registered users and 1 guest

cron

Remove Advertisements

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest